Mechanics and Construction > Mechanics and Construction

Robot horse neck


My 14yo daughter and I are building a robot horse.  I'm working on the mechanism to allow the neck to raise, lower, and move side to side.  My idea is to use 3 linear actuators in a triangle to create the motion.  the ends of the actuators will be free to swivel.  I am constraining the lower actuators to not point down (from the torso) and the upper actuator from not pointing up.   See the attached image of my first iteration of the mechanism.  (I'm new and couldn't figure out how to display the attached image inline)

Upon completing this first drawing, I worry that the weight of the neck might damage the stepper motors or the threaded rods due to high radial and axial loading.  (The neck and head frame weigh 3.9lbs bare and the head will eventually add 5-7 servos and a speaker, and it will eventually be covered in "fur".  The torque of just the frame is currently 3.3ft/lbs.  The center of the head and the weight of its future mechanisms is about 18" from the base of the neck, which adds an additional 3ft/lbs torque.  That means a static axial load of 8ish pounds on the top threaded rod and 4ish pounds on the bottom two.  Dynamic forces as the horse moves will be much higher.)  I am trying to figure out how to better support the weight of the front without hurting the steppers or the threaded rod.  I'd like the joint to be strong enough to prevent it from breaking the first time a child tries to pull down on the head.

Some ideas I've thought of (but am not fully sure of their value): 
[*]I can move the mounts on the torso further towards the top and bottom (while keeping the mounts on the neck in the same position) to trade off some radial load for more axial load.
[*]I can add a 608 bearing to the lead screws (mocked up on the bottom two steppers) to increase support.
[*]I could/ "hang" the neck with some tension springs to off-set the weight of the head/neck.  I don't have much experience making custom springs, but I realize they would need to have a certain tension so that they don't make the axial forces worse.

1)  Does anyone know the rough limitations of Nema 17 steppers as far as axial/radial load? 
2)  How much radial would 608 bearing to be able to handle?  (It obviously isn't a thrust bearing, but is designed for some axial forces.)  For cost reasons, I'd rather use 1 (or more) cheap 608 bearings than dedicated thrust bearings, if possible.
3)  Any suggestions on how best to bear the weight and torque of the neck/head in a mechanism like this? 
4)  Any major flaws of this mechanism that I'm not seeing?

I can't fully answer your questions, but maybe offer some tips and advice.  First off, NEMA 17 doesn't mean much except the size of the mount.
Within any mount size the electrical and mechanical aspects (power, speed, torque) can vary a lot.  So specifying NEMA 17 tells you pretty much nothing about its capabilities.  Second, if you don't need precise positioning, which I suspect you don't, I would recommend some kind of linkage instead of direct drive.  Cables of some sort would be great.  You can put the motors wherever works best and put loads on the motor bearings the way they were intended.  Third, springs would work but I think a counterbalancing weight would be better.  You can set it up in whatever way works best.  For instance, you can make it so the head is pretty well balanced or you can set it up so the counterweight pulls the head up until the motor pulls it down.  That way gravity is helping you in most cases.  Most steppers don't provide a lot of torque.  Cables can be attached in whatever position gives you the best strength vs speed compromise -- assuming the steppers have enough torque to move it at all.  The circuit you use to drive the steppers as well as the type of steppers you use will have influence on the torque and power available.
Hope this hekps


[0] Message Index

Go to full version